«DON’T PANIC: Tye’s intentionalist theory of consciousness* Alex Byrne, MIT Consciousness, Color, and Content is a significant contribution to our ...»
One might try to argue from these differences in representational vehicles to a difference in the kinds of contents represented. However, Tye does not supply any such argument, and there is no indication that he thinks one could be supplied. Moreover, it would be a confusion to think that a difference in representational vehicles entailed a difference in contents represented.
First, Heck’s version of the richness argument:
Consider your current perceptual state—and now imagine what a complete description of the way the world appears to you at this moment might be like.
Surely a thousand words would hardly begin to do the job. And it is not just that the description would be long: Rather, it seems hard to imagine that your perceptual state, as it is now, has any specific articulation corresponding to the conceptual articulation of a particular one of the many different Thoughts that might capture its content; and it seems at least as hard to imagine that you now possess all the concepts that would be expressed by the words occurring in such a description, even if one could be framed. Before me now, for example, are arranged various objects with various shapes and colors, of which, it might seem, I have no concept. My desk exhibits a whole host of shades of brown, for which I have no names. The speakers to the sides of my computer are not quite flat, but have curved faces; I could not begin to describe their shape in anything like adequate terms. The leaves on the trees outside my window are fluttering back and forth, randomly, as it seems to me, as the wind passes over them.—Yet my experience of these things represents them far more precisely than that, far more distinctively, it would seem, than any other characterization I could hope to formulate, for myself or for others, in terms of the concepts I presently possess.
The problem is not lack of time, but lack of descriptive resources, that is, lack of the appropriate concepts (2001, 489-90).
The conclusion of this argument is supposed to be that “the content of perceptual states is different in kind from that of cognitive states like belief” (485). Given the assumption (implicit in the quotation), that the content of belief is conceptual, the content view
Heck’s version of the richness argument overlaps with Tye’s: like Tye, he claims that experience represents, say, shades of color “of which, it might seem, I have no concept”. For example, one can have an experience of brown27, without having the concept brown27. And, as emphasized earlier, this does not have any tendency to show that perceptual experiences have a special kind of content.
However, the quoted passage contains another strand of argument, apparently leading to the conclusion that the content of perception cannot be fully expressed in any language—that perceptual content is not linguistic content. And if we add the premise that belief content can always be fully expressed in language, and further assume that linguistic content is conceptual, then the content view follows. So let us pursue this other strand for a moment.
The claim that perceptual content is not linguistic is not merely the claim that a particular perceiver might lack the vocabulary to express the content of his experiences.
This weak claim is no doubt true, but it evidently does not show that perceptual content resists expression in any language, and so does not show that the content of perception and the content of language are different in kind. Hence, Heck’s observation that his “desk exhibits a whole host of shades of brown, for which [he has] no names” does not support the claim that perceptual content is not linguistic: presumably the apparent shades of Heck’s desk can be captured linguistically with the aid of a paint catalogue. Rather, the crucial consideration is this: “it seems hard to imagine that your perceptual state, as it is now, has any specific articulation corresponding to the conceptual articulation of a particular one of the many different Thoughts that might capture its content” (clearly, given the context, we could replace ‘Thoughts’ by ‘sentences’). The idea here appears to
be the reverse of the official richness argument. It is not that perception is too finegrained to be captured by the net of language, but rather that language is too fine-grained:
to attempt to express perceptual content in language inevitably imposes on it a structure
experience gives it an unwanted conjunctive structure, and other unwanted structures would be introduced by any logically equivalent sentence (say, ‘~(~p v ~q)’).
This is certainly suggestive, but (at any rate in my brief exposition of the point) it is far too slender and elusive a reed to support any weight. Moreover, a similar point about belief would seem to be equally suggestive. Extruding beliefs through the templates of language often seems to impose on them unnecessary structure and precision. You realize you have forgotten your car keys, and so go back to the house to pick them up. The fact that you had some belief about the keys, together with an appropriate desire, explains your action. But what sentence expresses this belief? There are innumerable candidates: ‘I left the car keys on the kitchen table’; ‘I left the keys on the table in the kitchen’; ‘I forgot to pick up the keys from the table’; ‘The keys are where I left them, on the table’, etc. You are disposed to assent to all of these sentences, and so in this “dispositional” sense you believe the (different) propositions they express, but presumably not all of these beliefs causally explain your behavior. As Dennett puts it, “our linguistic environment is forever forcing us to give—or concede—precise verbal expression to convictions that lack the hard edges verbalization endows them with” (1981, 21). So, although Heck’s second strand of argument hints that perceptual content is not linguistic, a parallel strand hints that belief content is not (wholly) linguistic either.
And this is of course inconsistent with the content view. Nonetheless, I think Heck is onto something here; the issue is examined further in section 7.
Second, Peacocke’s argument for the content view:
Nonconceptual content has been recruited for many purposes. In my view the most fundamental reason—the one on which other reasons must rely if the conceptualist presses hard—lies in the need to describe correctly the overlap between human perception and that of some of the nonlinguistic animals. While
want to insist that the property of (say) representing a flat brown surface as being at a certain distance from one can be common to the perceptions of humans and of lower animals. The overlap of content is not just a matter of analogy, of mere quasi-subjectivity in the animal case. It is literally the same representational property that the two experiences possess, even if the human experience also has richer representational contents in addition. If the lower animals do not have states with conceptual content, but some of their perceptual states have contents in common with human perceptions, it follows that some perceptual representational content is nonconceptual (2001b, 613-4).32
This argument may be set out as follows:
1. Humans do, and the lower animals do not, “possess concepts”.
2. Humans are in states (e.g. beliefs) with conceptual content, and the lower animals are not.
3. Some of the perceptual states of the lower animals have contents in common with human perceptual states.
32 Cf. Bermúdez 1998, chs. 3 and 4; Evans 1982, 124; and McGinn 1989, 62. McDowell opposes this argument by denying premise (3): “We do not need to say that we have what mere animals have, nonconceptual content, and we have something else as well, since we can conceptualize that content and they cannot. Instead we can say that we have what mere animals have, perceptual sensitivity to features of our environment, but we have it in a special form. Our perceptual sensitivity to our environment is taken up into the ambit of the faculty of spontaneity, which is what distinguishes us from them” (1994, 64).
McDowell’s response is also endorsed by Brewer (1999, 177-9).
4. Human perceptual states have a kind of content that is not conceptual, i.e. they have nonconceptual content.
Since, by (2), human belief states have conceptual content:
5. The content view is true.
Because ‘possess concepts’ can be glossed in multiple ways, premise (1) can sustain a
variety of interpretations. It will be useful to distinguish three of them:
(1*) Humans have beliefs, and the lower animals do not.
(1**) Humans have beliefs with Fregean Thoughts as contents, and the lower animals do not have beliefs.
(1***) Humans have beliefs with Fregean Thoughts as contents, and the lower animals, although they may have beliefs, do not have beliefs with Fregean Thoughts as contents.
How does the argument fare on each of the three corresponding interpretations?
Not well on the first interpretation ((1) = (1*)). (1*) does not support the view that beliefs (unlike perceptions) have a special kind of content, and so does not support (2).
The second interpretation apparently conforms best to Peacocke’s intentions.33 It holds out more promise of supporting (2), but more needs to be said. On the face of it, 33 “I shall be taking it that conceptual content is content of a kind that can be the content of judgment and
one might reasonably hold (1**) together with the view that perceptual content, in humans and lower animals, is Fregean (i.e. conceptual)—thus denying (2).
On the third interpretation of (1), the lower animals might have beliefs with contents that are not conceptual. And, especially because the focus of the argument is on the overlap between humans and the lower animals, perhaps some human beliefs have such nonconceptual contents (why not?). So (1), on this interpretation, is in some tension with the conclusion of the argument, because the content view is at least committed to the claim that human belief exclusively has conceptual content. Further, the problem noted for the second interpretation also arises for the third.
Even if the problem for the second interpretation noted above can be overcome, there is the additional difficulty of justifying the claim that “the lower animals” (which Peacocke takes to include cats and dogs, and perhaps monkeys and apes34) enjoy perceptual experiences with contents in common with human perceptual experiences, while lacking beliefs. These issues are too large to be discussed here, but once it is conceded that having beliefs is not constitutively tied to speaking a language (as Peacocke himself is at pains to emphasize), then surely the burden of proof is on those who deny that humans and the lower animals have beliefs in the same robust sense.
Peacocke’s line of argument for the content view is an uphill struggle. What’s more, Tye himself would reject it completely. For according to him, fish have beliefs, and possess concepts (2000, 176-7). Notice that if fish lack beliefs, then none of their states 34 “Cats, dogs, and animals of many other species, as well as human infants, perceive the world, even though there conceptual repertoire is limited, and perhaps even nonexistent…the “soft line”…says that some of the conscious perceptual states with representational content enjoyed by mature humans can be enjoyed by nonlinguistic animals without concepts, or with only minimal conceptual capacities” (2001a, 260). And: “the soft line is right” (261). (I have ignored the hedging about “minimal conceptual capacities”.
It is absent in Peacocke’s 2001b, and so presumably Peacocke does not regard it as particularly significant.) 35 are poised: no state “stands ready and available” to affect beliefs.35 So, if fish lack beliefs, then the PANIC theory implies that there is nothing it’s like to be a guppy. Guppy consciousness is no doubt a bit fishy, but it is almost universally (and rightly) held that dogs and apes are phenomenally conscious. Hence, any reasonable PANIC theorist is committed to the view that these animals have beliefs, which puts him on a collision course with Peacocke’s “fundamental reason” for nonconceptual content.
To sum up the discussion of nonconceptual content. The PANIC theory interpreted on the state conception of nonconceptual content is inadequate (as Tye would no doubt agree). The right interpretation builds the content view into the PANIC theory.
However, we have found no reason to believe the content view: that beliefs have
content—nonconceptual content. Further, even if perceptual content is nonconceptual, Tye does not give any positive account of it. Lastly, because of the previous point, it is completely obscure why nonconceptual content (on the content conception) is part of ingredient X.
7 X=P+A+N revisited If the argument so far is correct, Tye has misidentified ingredient X: it is not P+A+N.
However, there are some important insights underlying his proposal—specifically the selection of P and N.
First, P. Its main role in the PANIC theory is to account for blindsight. In blindsight, the subject has a quasi-experience, say as of an ‘O’ before her, but (it is natural to say) she herself is unaware, or not conscious, that there is an ‘O’ before her.
35 “Standing ready and available” to affect desires is also sufficient for poisedness; but we may fairly
What I take to be the basic intentionalist insight about blindsight is this. The missing ingredient is not a non-intentional quale, or even a special kind of content, but simply the conscious subject herself. It does not seem to her that there is an ‘O’ before her. Assuming for simplicity that the content of her quasi-experience is the proposition that there is an ‘O’ before her, all that is required for phenomenal character is that it seems to the subject that there is an ‘O’ before her.